Finnish/Turkish: are they related?
Eugene Holman
Are Finnish and Turkish related?
REFORMULATING THE QUESTION
Finnish and Turkish show many superficial similarities, e.g. a system of
local cases and vowel harmony as in Fi omenassa Śin the appleą/kylässä Śin
the villageą, cf. Tu almada Śin the appleą/evde Śin the houseą; consonant
gradation and a genitive case in -n, e.g. Fi jalka Śfootą > jalan ŚGsgą/Tu
ayak > ayagąin Śidemą; use of a singular form of the noun with all
numerals, e.g. Fi kolme omenaa Śthree applesą/Tu üç alma Śidemą; a
preference for heavy premodification, e.g. Fi eilen kirjoittamani kirje
Śthe letter which I wrote yesterday (= yesterday written-my letter)/Tu dün
yazdigąim mektup Śidemą. Given the existence of these and many other
equally striking similarities, is there any justification for considering
Finnish and Turkish to have derived from a common earlier language?
BRIEF ANSWER
The methods used by historical linguistics do not allow this question to
be answered definitively. The similarities observed between Finnish and
Turkish warrant serious investigation. However, the lack of historical
records of earlier stages of these languages, our limited knowledge of the
nature of the historical contacts between the speech communities using the
languages from which Finnish and Turkish have developed, as well as the
limitations of the historical-comparative method do not yield a factual
output permitting us to regard these similarities as proof of genealogical
relationship.
DETAILED DISCUSSION
Similarities between languages can be accounted for in five ways:
1) Chance. Any two languages taken at random will show several dozen
similar words and structures , e.g. Indonesian Śduaą and Latin Śduaeą =
Śtwoą, Swedish Śi bankeną and Bulgarian Śv bankataą = Śin the banką, both
with the structure preposition + (NOUN + definite article).
2) Linguistic universals. Human languages all share certain features of
design, given the constraints posed by human anatomy and perception. For
example, languages must dispose over some means for indicating spatial
relationships, English and Indonesian put the relator before the head
word, e.g. Śin the banką, Śdi banką, Finnish and Japanese put it after it:
Śpankissaą, Śginkoo nią.
3) Proximity. The similarities in the sound systems of Finnish and
Finland-Swedish such as lack of aspirated voiceless plosives or the
drop-rise accent (accent 2), for example, are attributable to their having
been spoken in contiguous or overlapping speech communities which have
been interacting with each another for almost a millennium
4) Cultural contact. The presence in Japanese of the word Śbiiruą (<
Dutch bier Śbeerą) is known to derive from cultural contacts which can be
located with respect to their time, place, and direction.
5) Genealogical relationship. By this we mean that the divergent forms
observed in two different languages at a specific point in time (b) can be
demonstrated to be different developments of a single form which existed
in a single language which was spoken at an earlier time (a). Thus Spanish
Śochoą and French Śhuitą can both be regarded as different modern versions
of the Latin form Śoctoą - Śeightą.
If the similarities observed between Finnish and Turkish could be shown to
be different developments of single forms which existed in a single
language spoken at an earlier time, this would demonstrate the existence
of a genealogical relationship between them.
GENEALOGICAL RELATIONSHIP
The method used to demonstrate genealogical relationship requires
various types of information as its input. The absolute minimum is
accurate descriptions of the languages to be compared. Written records and
other information shedding light on earlier forms of the languages being
compared and the historical development of the speech communities using
them can also help filter out Śnoiseą: similarities attributable to
non-genealogical factors. As the comparison of Spanish Śochoą and French
Śhuitą show, genealogical relationship is not always immediately obvious,
several intermediate stages have to be posited before the relationship
linking Śochoą and Śhuitą to Latin Śoctoą can be established. These
intermediate stages are not arbitrary, but rather must be consistent with
our picture of how the languages in question have developed in general.
Without going into details, we can say that the equation Sp Śochoą & Fr
Śhuitą < Lat Śoctoą is based upon sound changes which are regular enough
to enable us understand the equation Sp Ślecheą & Fr Ślaitą < Lat Ślactaą,
and more abstractly, Sp -ch- & Fr (-t) < Lat -ct-. They can help us
interpret and evaluate further data, e.g. Sp Śhechoą, Fr Śfaitą < Lat
Śfactu-ą; Romanian Ślaptaą, Śoptą. Śfaptą. It should be obvious from these
examples that genealogical relationship does not imply similarity, nor
does similarity imply genealogical relationship.
THE RELATIONSHIP BETWEEN FINNISH AND TURKISH
The possibility of a genealogical relationship between the Uralic
languages, to which Finnish belongs, and the Altaic languages, to which
Turkish belongs, was first hypothesized by the Danish scholar Rasmus Rask
(1787-1832), one of the founders of modern scientific linguistics, on the
basis of first-hand contacts he made with the languages of the interior of
Russia. More detailed data for the comparison of the Uralic and Turkic
languages was gathered by the Finn M. A. Castren (1813-1852). A
systematization of the lexical and grammatical correspondences was made by
the German Wilhelm Schott (1802-1889), and he is the first one to have
actually claimed the existence of a genealogical relationship between the
Uralic and Altaic languages.
Scholars working in this area who have presented evidence and arguments
supporting the hypothesis of a relationship between Uralic and Altaic
include A. Sauvageot, B. Serebrennikov, and B. Collinder. Those who have
argued against it include G. I. Ramstedt, E. Itkonen, and D. Sinor. The
American Eurasianist Robert. Austerlitz (1923-1994), while not going so
far as to commit himself as a supporter of the hypothesis, emphasized the
importance of studying in greater detail the vast northern Eurasian
linguistic landscape where agglutinating languages exhibiting a complex
network of structural and historically motivated (contact, proximity,
common origin) similarities are indigenous. This area includes Finnish,
Saami, and Turkish as well as Mongolian, Korean, Japanese, and Gilyak. The
question of how these similarities are to be interpreted cannot be
answered conclusively until the relationships between the Altaic languages
are better understood. The understanding of their historical development
thus obtained would allow for the reconstruction of proto-Altaic and the
possible construction and evaluation of Altaic-Uralic and
Altaic-Finno-Ugric etymologies.
THE SIMILARITIES BETWEEN FINNISH AND TURKISH
The similarities between Finnish and Turkish include:
PHONOLOGY
1) vowel harmony, 2) consonant gradation, 3) a simple syllable structure
which avoids consonant clusters at word margins.
MORPHOLOGY
4) no grammatical gender, 5) no articles, 6) synthetic agglutination as
the favored grammatical technique, 7) possessive suffixes, 8) an abundance
of verbal nouns, adjectives, and adverbs, 9) a genitive case in -n, 10) a
system of local cases based on a tripartite distinction between Ścomingą,
Śbeingą, and Śgoingą, 11) a separative case with comparison, 12) the
definite/indefinite opposition indicated by opposition of two noun cases,
13) singular form of nouns after numerals higher than one.
SYNTAX
14) postpositions rather than prepositions, 15) lack of a verb meaning Śto
haveą, 16) use of a negative verb, 17) use of a particle as an
interrogative marker, 18) preference for pre-modification.
If what we know about changes which have taken place in Baltic-Finnic as
a consequence of contact with Indo-European languages is admitted as
additional evidence, we can add:
19) lack of agreement between adjectives and the nouns they modify, 20) a
preference for SOV word order
There is a notable lack of similarity in basic vocabulary.
INTERPRETING THE SIMILARITIES
As long as the list of similarities is, it is not sufficient to
demonstrate relationship. Rather, the features in question can be
attributed to a combination of chance factors and linguistic universals.
Languages which use synthetic agglutination as their favored grammatical
technique will tend to develop organizational strategies which facilitate
the construction and processing of the lengthy and grammatically complex
words which this necessarily presupposes. Finnish and Turkish appear to
have so many Śunusualą structural features in common because most of us
are used to looking at analytical western European languages which
themselves share many organizational strategies but are structurally quite
different from Finnish and Turkish. As impressive as the list is, then,
similar organizational strategies do not constitute proof of genealogical
relationship. English and Chinese, both analytical languages with
rudimentary inflectional systems, share many organizational strategies but
are not (as far as we know) related; analytic isolating Modern English and
synthetic inflecting Old English share few organizational strategies but
they are definitely related.
WHAT KIND OF SIMILARITY PROVES GENEALOGICAL RELATIONSHIP?
Historical linguistics requires that the grammatical and lexical core of
the languages being compared exhibit systematic correspondences in form
and meaning. Information about the specific developments undergone by the
languages in question as well as about the external history of the speech
communities concerned facilitate the interpretation of the material to be
compared. If we did not know that Finnish word final -si derives from an
earlier word final -te the relationship between Finnish vesi Śwaterą, käsi
Śhandą and their Erzya Mordvin cognates vedą, kedą would be more difficult
to formulate. Similarly, if we were ignorant of the fact that Afrikaans
developed from Netherlandic dialects transplanted to Africa during the
17th century we would have difficulties explaining the similarities and
differences observed when comparing it to Zulu and modern Dutch.
The methodology of historical linguistics enables us to trace the basic
grammatical and lexical core of Finnish to proto-Uralic, a language
thought to have been spoken approx. 4000 BC, probably in the area north of
the forest zone-steppe border between the Volga and the Ob Rivers. This
language, presumably the speech of a Neolithic community of hunters and
gatherers who were experimenting with agriculture and fixed settlements,
was itself the product of historical evolution, about which we know
nothing. The information yielded by historical linguistics about the
speech, location, and dating of the proto-Uralic speech community reveals
nothing about the ultimate origin of Finnish and the langages related to
it. It is nothing more than the oldest reliable information about a
language with a complex structure and vocabulary which themselves had
evolved under conditions inaccessible to current methods of historical
research.
The basic grammatical and lexical core of Turkish has developed from
languages, the history of which is only sketchily known, but which points
to somewhere in Siberia. The manner in which the Turkic languages
(Turkish, Uzbek, Turkmenian, Tatar, Kirghiz, Azeri, Chuvash, Yakut,
Bashkir, etc.) are related both to each other and to the other Altaic
languages (Mongolian, Tungus, according to some scholars also Korean and
Japanese) is by no means clear, and many questions concerning their
historical evolution await clarificaton. In contrast to the relative
unanimity which reigns concerning the reconstruction of proto-Uralic,
scholars have only succeeded in reconstructing fragments of proto-Altaic,
with many specialists doubting whether a proto-Altaic language or speech
community ever existed.
The methods used to demonstrate genealogical relationship were
elaborated using material from languages with well documented internal and
external histories. They assume the existence of relatively stable
populations, definite areas of settlement, and, ideally, continuity of
written records documenting earlier forms of the language. The population
of the areas where the earliest forms of Uralic and Altaic languages are
thought to have been spoken is known to have led a nomadic life and to
have been intermingling with each other for millennia, nor did they leave
written records. Nothing that we know excludes the possibility of there
having been various types of cultural contacts between the proto-Uralic
and earlier Altaic speech communities, nor can we preclude the possibility
of these speech communities having been neighbors, shared the same
territory, or even experienced various periods of Uralic-Altaic
bilingualism. At least one Turkic language, Chuvash, has an obvious
Finno-Ugric substratum, and borrowings from Turkic, sometimes on a massive
scale and indicating anything ranging from centuries to millennia of
cultural contact, are found in Mari, Mordvin, Komi, Udmurt, Mansi, Ostyak,
the Samoyed languages, and, of course, Hungarian. How far back these date,
and how they are to be interpreted has not yet been fully worked out by
scholarship.
The Swedish Uralicist Björn Collinder, one of the foremost proponents of
an ancient genealogical relationship between Altaic and Uralic, focused
attention on the existence of what he considered to be a small number of
cognates - words the form and content of which derive from a common
source, in Uralic or Finno-Ugric, and Altaic. The forms he posited as
possible Uralic/Altaic cognates include: Fi ala Śunderą/Tu al Śsideą Fi
kaalaa- Śto wadeą/Tu kel- Śto comeą, Fi ken- Śwhoą/Tu kim Śwhoą, Fi kuule-
Śhearą/Tu kulak Śearą, Fi mi- Śwhatą/Tu -mi Śwhatą, Fi minä ŚIą/OldTu bän,
män, min ŚIą, Fi saarna Śsolomn utteranceą/Tu saryn Śsongą, Fi unohta- Śto
forgetą/Turk unut- Śto forgetą. The forms he posited as possible
Finno-Ugric/Altaic cognates include: Fi koi Śmothą/Tu küjä Śmothą, Fi kylä
Śvillageą/(Volga) Turkish kül, kil Śyardą, syylä (<*c´üklä) Śwartą/Tu
sigil Śwartą, tyvi (*tünge) Śbutt, stumpą/Tu tünggäk Śstump of treeą (B.
Collinder: Fenno-Ugric Vocabulary, 1977, pp. 152 ff.)
Whether such items can be used as proof of a relationship between the
Uralic and Altaic families, and thus between Finnish and Turkish, is
uncertain. They cannot be overlooked, although the degree to which we are
dealing with similarities resulting from chance, proximity, or cultural
borrowing is unclear, given the limitations posed by our methodology, the
lacunae in our knowledge of northern Eurasian prehistory, and the time
depth involved.
WHAT ABOUT FINNISH AND HUNGARIAN?
At first glance Finnish and Hungarian (here in a quasi-phonological
transcription) neither look nor sound the same: e.g. Hyvää päivää/Joo
napot ŚGood dayą, Puhut hyvin englantia. Jool angolul beseels. ŚYou speak
English wellą, Anteeksi, missä ravintola on?/Bo˘aanotot, hol van az
eetterem? ŚExcuse me, where is the restaurant? Rakastan sinua/Seretlek ŚI
love youą. Closer inspection reveals numerous superficial structural
similarities, e.g. local cases and vowel harmony as in Fi omenassa Śin the
appleą/kylässä Śin the villageą, cf. Hu az almaaban Śin the appleą/a
keezben Śin the handą; use of a singular form of the noun with all
numerals, e.g. Fi kolme kalaa Śthree fishą/Hu haarom hal Śidemą; a
preference for heavy premodification, e.g. kaupungissa oleva ystäväni Śmy
friend, who is in the city (= in-(the)-city being friend-my)ą/a vaaroshban
valoo baraatom Śidemą. More importantly, however, numerous items in the
basic vocabulary of the two languages bear obvious resemblances to each
other in form and meaning, e.g. kala/hal Śfishą, mi-/mi Śwhatą, anta-/ad-
Śto giveą, silmä/sem Śeyeą, veri/veer Śbloodą, käsi/keez Śhandą,
pilvi/felhöö Ścloudą, puno-/fon- Śto twistą, alla/alott Śunderą etc. Given
the existence of these and many other equally striking similarities, how
can we evaluate the relationship between Finnish and Hungarian?
BRIEF ANSWER
Strict application of the historical-comparative method supplemented by
information gathered from historical records documenting earlier stages of
these and other languages, yield a factual output permitting us to regard
a subset of the similarities observed as indisputable proof of
genealogical relationship.
DISCUSSION
When seeking to determine whether two languages are genealogical related
we do not focus on superficial similarities between words or grammatical
structure. The former could be the product of chance or cultural
borrowing, as witnessed by the presence of the word hanbaagaa Śhamburgerą
in Japanese, the latter could be the result of constraints dictated by the
limited number of structural patterns and techniques used by human
languages. What we are searching for is recurrent patterns of sound
correspondence between words of the basic and historically oldest layer of
vocabulary which show similarities in form and meaning.
The comparison of words from different languages is not a simple
procedure. Before the comparison is made, the linguist tries to ensure
that the phonetically oldest possible form of the word is being compared.
This often requires the use of forms from ancient documents and
non-standard dialects if they are available and the relationship of the
forms in question to those of the standard are known. Even if such forms
are not available, the linguist will often find it necessary to perform
internal reconstruction before attempting the actual comparison.
Let us consider the Finnish word käsi Śhandą. The dictionary form
contains the intersyllabic consonant -s- and the word final vowel -i, but
many of the inflected forms contain -t- or -d-, followed by -e, e.g.
käteen Śinto the handą, kädessä Śin the handą. Our knowledge of the
manner in which Finnish has evolved tells us that the -s- and -d- are the
results of known sound changes affecting older *-t-; the forms in -i are
the results of known sound changes affecting word-final *-e. Thus, when
comparing the Finnish word with words in other languages we should use the
reconstructed oldest form käte- rather than the misleading dictionary form
käsi. Insofar as possible, the forms from all the languages to be compared
should first be subjected to internal reconstruction.
Let us now consider some basic Finnish lexical items beginning with the
sound p-, and their cognates in Erzya Mordvin, Komi, and Hungarian. Forms
which are the result of internal reconstruction are folowed by a hyphen.
The forms in Erzya Mordvin, Komi, and Hungarian are given in a
phonological transcription:
Finnish Erzya Mordvin Komi Hungarian Gloss
1. pel-k- pele- pol feel fear
2. pesä pize poz feese- nest
3. pilve- pelą piv fel- cloud
4. pitä- pedąa - füüz- to keep, bind
5. puno- pona- pyn- fon- to twist
6. puu - pu fa tree
7. pää pe pon fej head
Certainly the basic criteria, that there be systematic correspondences
between form and meaning, and that on the level of form there be
systematic sound correspondences, seem to have been met. Finnish-Erzya
Mordvin-Komi word-initial p- seems to correspond regularly to Hungarian
word-initial f-.
Let us look at some more cognates:
Finnish Erzya Mordvin Komi Hungarian Gloss
8. kala kal - hal fish
9. kivi kev - köve- stone
10. kolme kolmo kujim haarma- three
11. kota kudo kola haaz house, hut
12. ku- ko- ko- ho- wh(ere), wh(ich)
13. kuute- koto kvatą hat six
14. käte- kedą ki keze- hand
Here, the k- of the other languages seems to correspond to two Hungarian
sounds, h- and k-. Closer inspection, however, reveals, that when Finnish
and the other languages have a front vowel following the k-, the result in
Hungarian is k-, otherwise we have the expected h-. I use the word
Śexpectedą because the relationship between p and f is repeated in the
relationship between k and h: in both instances we have a relationship
between stops and fricatives which share approximately the same place of
articulation. A relationship of this complexity would, paraphrasing Sir
William Jones, be difficult to account for in any other manner except
hypothesizing common relationship.
Let us look at some more sets of cognates:
Finnish Erzya Mordvin Komi Hungarian Gloss
15. anta- ando- ud ad- to give
16. mene - mun- men- to go
17. mi- me- myj mi what?
18. miniä - moną meną- sister-in-law
19. muna mona - mony egg, testicle
20. neljä nile nąolą needą four
21. niele- nile- nąyl- nąel- to swallow
22. nuole- nal nąil nąila arrow
23. pala- palo- - fadą to freeze, burn
24. sarve- sąuro sąur sarv horn
25. sata sąado sąo saaz hundred
26. silmä seląme sąin sem eye
27. suone- san sőn iin sinew
28. sydäme- sedej sąőlőm sive- heart
29. syksy soksą - öös autumn
30. syle- sëlą syl öl bosom
31. talve- tele től tele- winter
32. uute- od vylą uuj new
33. vere- verą vir vere- blood
34. vete- vedą vit vize- water
35. viite- vedąa vit öt five
Interpreting the relationships between the words requires detailed
information about the individual history of the languages concerned. For
example, Finnish viisi and Hungarian öt Śfiveą do not at first glance seem
to have anything in common. But when we replace viisi with viite-, the
product of our internal reconstruction, and see that older documents in
Hungarian indicate that öt once began in a consonant v-, and that an older
vowel e was labialized to ö under its influence before it was lost, the
relationship between Finnish viite- and an older Hungarian *wet becomes
clearer.
Comparison between Finnish, Hungarian, and the other Uralic languages
focusing on their basic vocabulary and inflectional elements reveals the
existence of dozens of systematic sound correspondences of the type
illustrated above. This information, supplemented by the study of
loan-word strata, the individual historical evolution of the two languages
and of the speech communities using them, provides a solid basis for
concluding that common origin rather than chance, proximity, or cultural
contact most plausible accounts for the similarities observed between
Finnish and Hungarian.
Eugene Holman
University of Helsinki
Okan Ertek
Kiitos.
Would you please re-post the same article but this time in Finnish
please.
Thanks,
OKAN ERTEK
Mighan Sung
Japanese belong to Uraic-Altaic language group.
They believe these people originated from the central Asia and dispersed
throughout these countries.
Eugene Holman
The purpose of my posting was to demonstrate that this cannot be
demonstrated using the methods of historical and comparative linguistics.
Finnish, Estonian, Sami (Lappish), Mordvin, Mari, Udmurt, Komi, Khanty,
Mansi and Hungarian belong to the Finno-Ugric family of the Uralic
macro-phylum, which also includes the Samoyed languages Enets, Nenets, and
Nganassan. Many scholars consider it to be linked to Yukaghir as well.
Turkish, along with Azeri, Uzbek, Chuvash, Turkmenian, Kazakh and several
other languages of central Asia form the Turkic family. Their relationship
is easily demonstrated by linguistic means. More complex is the notion of
an 'Altaic' family, which would include Mongolian, Tunguz, and possibly
Korean. Korean and Japanese show many similar traits, but the extent to
which these are the result of mutual influence or borrowing is difficult
to determine, and many scholars consider Japanese to be a linguistic
isolate.
The 19th century notion of a Ural-Altaic family embracing Finnish,
Hungarian, Turkish, Mongolian, Korean, and Japanese was based primarily on
shared organizational features such as agglutination, vowel harmony, and
lack of grammatical gender. The more rigorous methods of modern
comparative linguistics require cognates sharing phonological
correspondences as a demonstration of common descent. These can be found
for Finnish, Hungarian, and Samoyed, but they cannot be found between
Uralic, Turkic, and the other languages mentioned, hence no demonstration
of relationship. Note, I'm not denying the possibility of a relationship,
I'm just saying that the methods presently used to demonstrate
relationship do not yield a positive result. See Anthony FOX, *Linguistic
Reconstruction: an Introduction to Theory and Method*, Oxford Univ. Press,
1995, for a serious discussion of the methodological issues involved, and
Merrit RUHLEN, * Guide to the World's Languages, vol. 1, Classification*,
Stanford University Press, 1991, particularly pgs. 64 - 71, for a
discussion of the history of the Ural-Altaic hypothesis.
Regards,
Eugene Holman
Richard M. Alderson III
hol...@katk.helsinki.fi (Eugene Holman) writes:
>Turkish, along with Azeri, Uzbek, Chuvash, Turkmenian, Kazakh and several
>other languages of central Asia form the Turkic family. Their relationship is
>easily demonstrated by linguistic means. More complex is the notion of an
>'Altaic' family, which would include Mongolian, Tunguz, and possibly
>Korean. Korean and Japanese show many similar traits, but the extent to which
>these are the result of mutual influence or borrowing is difficult to
>determine, and many scholars consider Japanese to be a linguistic isolate.
The Altaic family was originally thought to comprise Turkic, Mongolic, and
Manchu-Tungusic. Although there is apparently a strongly politically motivated
group in Europe who deny the relationship (cf. the paper [by Miller, I think]
on the issues in _Sprung from Some Common Source_, ed. Lamb), the evidence when
combined with Occam's Razor is too strong to deny.
There have been more recent researchers who accept the Altaic family as such
who have then tried to connect Korean or Japanese thereto, each independent of
the other language. If Martin's connection of Japanese and Korean is correct,
then if either can be shown to be related to Altaic (as either a sister or a
daughter), the other gets a free ride; on the other hand, both independently
may be related to Altaic, and Martin's work might then show a close subgrouping
without necessarily a common protolanguage.
There is also the issue of an Austric substrate in Japanese, the evidence for
which I have not yet looked for, since I don't usually work in that part of the
world.
--
Rich Alderson You know the sort of thing that you can find in any dictionary
of a strange language, and which so excites the amateur philo-
logists, itching to derive one tongue from another that they
know better: a word that is nearly the same in form and meaning
as the corresponding word in English, or Latin, or Hebrew, or
what not.
--J. R. R. Tolkien,
alde...@netcom.com _The Notion Club Papers_
Michael Ernst
I thought that the Uralic languages were divided into two (sometimes
three) subdivisions: Fenno-Ugric (or sometimes Fennic and Ugric) and
Samoyeds - the first group containing Finnish, Estonian, Hungarian and
the Sami languages; and the second group being comprised of a small
group of languages spoken in various regions of Siberia.
I would like to also ask what exactly the Aryan languages are. All the
sources that I've seen give a very vague answer.
-Michael J. Ernst
Eugene Holman
(Michael Ernst) wrote:
> I thought that the Uralic languages were divided into two (sometimes
> three) subdivisions: Fenno-Ugric (or sometimes Fennic and Ugric) and
> Samoyeds - the first group containing Finnish, Estonian, Hungarian and
> the Sami languages; and the second group being comprised of a small
> group of languages spoken in various regions of Siberia.
No, you have it all wrong.
The Uralic macro-phylum of languages consists of two related families of
languages: Finno-Ugric and Samoyed.
The Finno-Ugric languages are subdivided into the following branches:
1. Finnic:
a) Baltic-Finnic:
- Finnish: 5,000,000
- Estonian
- Karelian, including Olonetsian and Ludian: 75,000
- Vepsian: 6,000
- Livonian: ?50
- Ingrian: 300
- Votian: ?50
b) Sami (Lapp):
- Northern Sami: 30,000
- Inari Sami: 400
- Skolt Sami: 300
- Kildin Sami: 800
- Southern Sami: ?500
- Lule Sami: ?2,000
c) Mordvin
- Erzya: 500,000
- Moksha: 250,000
d) Mari
- North-Western Mari: 10,000
- Hill Mari: 30,000
- Meadow Mari: 500,000
e) Permic languages:
- Udmurt: 500,000
- Permyak: 100,000
- Komi: 250,000
2. Ugric:
a) Ob-Ugric:
- Mansi: 3,000
- Khanty: 13,000
b) Hungarian: 14,000,000
The Samoyed languages are:
a) Nganasan: 600
b) Enets: 50
c) Nenets: 27,000
d) Selkup: 1,500
The information on number of speakers was optained from the Department of
Finno-Ugric Studies, University of Helsinki, webpage, accessible from:
URL: http://amacrine.berkeley.edu/finnugr/uralic-table.html.
I have used slightly differnet terms for some of the ones used there, but
the differences are insignificant.
The main controversy surrounding this classification concerns the
relationship between Mordvin and Mari. Some scholars group them together
under a single Volga-Finnic mode, others are of the opinion that the
similarities between them do not support the postulate of their having
developed from A single Volga-Finnic proto-langauge.
For a discussion of the evolution of views concering the classification of
the Uralic languages see: Merritt RUHLEN, 1991, *A Guide to the World's
Languages, volume 1: Classification.*, Stanford University Press,
Stanford, ISBN 0-8047-1894-6, Library of Congress Classification P203.R8
1991, page 64 ff.
>
> I would like to also ask what exactly the Aryan languages are. All the
> sources that I've seen give a very vague answer.
>
The term *Aryan* was once (19th century) used as a cover term for the
languages descended from the easternmost branch of the Indo-European
languages: the branches that are today referred to as Iranian and Indic.
When a cover term is needed today, Indo-Iranian is used.
The term Aryan fell into disrepute due to Nazi ratial politics and
ideology. They developed a mythical image of Aryans as tall, blond-haired,
blue eyed warriers. In actual fact, the peoples the term refers to when
used 'properly' include the ancestors of the Afghanis, Persians, and
Punjabis.
Regards,
Eugene Holman